At the launch of Quantum Marketing, we had the pleasure of welcoming Ariane Lambert Mongiliansky who promised us the imminent release of her research paper on quantum persuasion. This is now a reality as the Quantum Interaction 2018 conference unveiled the secrets of this new approach that demonstrates that distraction is a powerful persuasive factor.

This experiment was conducted jointly with the help of Hervé GONAY, founder of GetQuanty and Adrien Calmettes from the University of Nice.

The theory of persuasion was initiated by Kamenica and Gentskow and has developed in various directions. The purpose of persuasion theory is the use of an information structure (or measure) that generates new information in order to modify a person's state of belief with the intention of making them act in a specific way. The question of interest is to know to what extent a person, calling him Sender, can influence another, calling his receiver, by selecting an appropriate measure and revealing his result. One example is lobbying. A pharmaceutical company commissions a scientific laboratory to carry out a specific study of the impact of a drug, the result of which is given to the politician. The question of interest from the point of view of persuasion is to know what type of study best serves the interests of the company. The recipient's decision to act depends on his or her beliefs about the world. Beliefs are given in the form of a distribution of probabilities over a set of states in the world. A central assumption is that uncertainty is formulated within the standard classical framework. Consequently, the updating of the recipient's beliefs follows Bayes' rule. However, as widely documented, the functioning of the mind is more complex and often people do not follow Bayes' rule. Cognitive sciences offer alternatives to Bayesian studies. One of the avenues of research in cognitive sciences uses the formalism of quantum mechanics. One of the main reasons is that quality management has properties that recall the paradoxical phenomena of human cognition. There are also deeper reasons to turn to quantum mechanics when studying human behaviour. In addition, cognition has been able to explain a wide variety of behavioural phenomena such as disjunction effect, cognitive dissonance or preference inversion. Finally, there is now a fully developed decision theory based on the principle of quantum cognition. It is clear that the mind is likely to be even more complex than a quantum system, but we believe that the quantum cognitive approach already offers interesting new perspectives, especially with regard to persuasion.

In quantum cognition, the system of interest is the mental representation of the world by the decision-maker. It is represented by a cognitive state. This representation of the world is modelled as a quantum system, so that the uncertainty of the decision is of a non-conventional (and therefore quantum) nature. As we argue, this modelling approach captures the generalized cognitive difficulties that people present when constructing a mental representation of a "complex" alternative. The key quantum property we use is that some characteristics (cf. properties) of a complex mental object can be "complementary to Bohr" which is incompatible in the mind of the decision-maker: it cannot combine information from both points of view in a stable way. A central implication is that measures (new information) modify the cognitive state in a well-defined and non-Bayesian way. As in the traditional context, our rational recipient uses new information to update his or her beliefs so that choices based on the updated preferences are consistent with the ex ante preferences defined for the condition (event) that triggered the update. We have learned that a dynamically coherent rational and rational decision-maker updates her beliefs according to von Neumann's postulate. In two recent articles, important theoretical results have been established. First, it is demonstrated that in the absence of any constraint on the measures, full persuasion applies: the sender can always persuade the recipient to believe everything that is favourable to him. Then, the same authors study a short sequence of measures but in a simpler task that we call "targeting". The purpose of "targeting" is the transition from one state of belief to another specified target state. The main findings relevant to our question are that the distraction of "providing" irrelevant or "incompatible" information has a significant persuasive power. In a Bayesian context, this information should have little or no effect.